A method, device, computing device, and system for monitoring the performance of parallel water pumps

Abstract

The present invention discloses a method, device, computing device and system for monitoring the performance of parallel pumps. According to the technical solution provided by the present invention, parallel circulating pumps are monitored, and historical operation data of the parallel circulating pumps is collected; data cleaning is performed on the collected historical operation data to obtain repaired historical data after repair; wherein, data cleaning includes: abnormal data detection, abnormal data screening and / or data repair; based on the repaired historical data, a pump characteristic curve is generated, and a performance model of the parallel circulating pump is constructed; based on the current operation data of the parallel circulating pump, the operation state of the parallel circulating pump is monitored and optimized by using the performance model of the parallel circulating pump. Through the present invention, based on the historical operation data of the parallel pumps, a performance model of the parallel circulating pump is generated by fitting the performance curve, so as to realize the real-time monitoring and analysis of the pump performance, scientifically predict the operation condition of the pump, and obtain a reasonable optimization method.

Description

Technical Field

[0001] The present invention relates to the field of industrial power generation, and particularly to a method and device for monitoring the performance of parallel pumps, a computing device, and a system. Background Art

[0002] The thermal power industry is the main power source for power supply in China. As one of the important auxiliary machines in thermal power plants and heat power plants, the power consumption of circulating pumps accounts for a relatively high proportion in the self - use power rate of heat power plants. Therefore, real - time monitoring of the operating performance of circulating pumps and reasonably using the start - stop of circulating pumps to ensure their operation in the high - efficiency area are important measures for energy conservation and consumption reduction in heat power plants.

[0003] In this regard, the existing circulating pump monitoring technologies mostly rely on traditional sensors and control systems, lacking in - depth analysis and optimization of pump performance. These technologies usually only focus on the real - time response of heat supply and demand, while ignoring how the operating performance of each pump is during the response process. This makes the operating data of each pump obtained usually relatively single, and the data collection is not comprehensive, resulting in the inability to calculate the pump efficiency based on the operating parameters and the failure to achieve real - time monitoring of the pump operating performance. Moreover, in actual work, due to factors such as site limitations and connection installations, the installation layouts of parallel circulating pumps are all different. As a result, during their operation, even for pumps of the same model, there will be obvious differences in operating parameters under the same working conditions. This difference causes some pumps to have a large operating pressure, while the performance of some other pumps cannot be fully utilized, making the overall working efficiency of parallel circulating pumps relatively low and increasing their overall power consumption. Summary of the Invention

[0004] To solve the above - mentioned technical problems, the present invention provides a method for monitoring the performance of parallel pumps, and a corresponding device, computing device, and system for monitoring the performance of parallel pumps.

[0005] According to one aspect of the present invention, there is provided a method for monitoring the performance of parallel pumps, the method comprising:

[0006] Monitoring parallel circulating pumps and collecting historical operating data of the parallel circulating pumps;

[0007] Performing data cleaning on the collected historical operating data to obtain repaired historical data; wherein, data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair;

[0008] Generating a pump characteristic curve based on the repaired historical data and constructing a performance model of the parallel circulating pumps;

[0009] Based on the current operating data of the parallel circulating pumps, using the performance model of the parallel circulating pumps to monitor and optimize the operating state of the parallel circulating pumps.

[0010] In the above solution, monitoring the parallel circulating water pumps and collecting the historical operation data of the parallel circulating water pumps further includes:

[0011] The historical operation data at least includes the inlet pressure of the main pipe, the outlet pressure of each pump, the total amount of circulating water, the opening of each hydraulic coupling scoop tube, and / or the current of each motor.

[0012] In the above solution, cleaning the collected historical operation data to obtain the repaired historical data; among them, data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair, and further includes:

[0013] Generating structured monitoring data of the circulating water pump based on the historical operation data with the same sampling frequency;

[0014] Using the 3σ criterion to detect abnormal data for the structured monitoring data;

[0015] Based on the abnormal data detection results, screening to obtain an abnormal data set;

[0016] Based on the LSTM neural network and the SVR algorithm, performing data repair on the abnormal data set;

[0017] The data repair at least includes: redundant data elimination, incorrect data correction, and / or missing data filling.

[0018] In the above solution, generating the characteristic curve of the water pump and constructing the performance model of the parallel circulating water pump based on the repaired historical data further includes:

[0019] Based on the repaired historical data and the structural parameters of the parallel circulating water pump, determining the local resistance coefficient, and based on the dynamic head and the static head, determining the head of the parallel circulating water pump;

[0020] Based on the head of the parallel circulating water pump, determining the corresponding flow rate, input power, and rotational speed, and generating the characteristic curve of the parallel circulating water pump; among them, the characteristic curve at least includes the Q-H curve, the Q-P curve, and the Q-η curve;

[0021] Establishing a water pump efficiency model, a water pump power model, a motor efficiency model, and a hydraulic coupling efficiency model;

[0022] Performing polynomial fitting on each characteristic curve of the parallel circulating water pump, and accordingly constructing the performance model of the parallel circulating water pump.

[0023] In the above solution, based on the current operation data of the parallel circulating water pump, using the performance model of the parallel circulating water pump to monitor and optimize the operation state of the parallel circulating water pump further includes:

[0024] Obtain the current operating data of the parallel circulating pump and determine the actual performance data;

[0025] Input the current operating data of the parallel circulating pump into the performance model of the parallel circulating pump, and calculate the predicted performance data;

[0026] Compare the actual performance data with the predicted performance data to obtain a comparison result;

[0027] Adjust the operating state of the parallel circulating pump according to the comparison result.

[0028] In the above solution, the step of comparing the actual performance data with the predicted performance data to obtain a comparison result further includes:

[0029] Calculate a corresponding preset number of predicted performance data based on a preset number of repair history data as the initial data set;

[0030] Perform data filtering on the initial data set, and obtain a filtered data set after completion;

[0031] Calculate the upper limit parameter and the lower limit parameter based on the filtered data set;

[0032] Based on the upper limit parameter and the lower limit parameter, determine the threshold range for comparing the actual performance data with the predicted performance data.

[0033] In the above solution, the method further includes:

[0034] Based on the initial data set, calculate the average value of all the actual performance data therein , as well as the upper quartile and the lower quartile , and accordingly determine the filtering upper limit and the filtering lower limit

[0035]

[0036]

[0037] According to the filtering upper limit and the filtering lower limit filter the initial data set to obtain a filtered data set, and calculate the average value of the filtered data set

[0038]

[0039] Among them, is the number of actual performance data in the filtered data set, is the actual performance data in the filtered data set;

[0040] Upper limit parameter and lower limit parameter are

[0041]

[0042]

[0043] wherein, is the number of data in the filtered dataset that is greater than the average value of the filtered dataset ; is the data in the filtered dataset that is greater than the average value of the filtered dataset ; is the number of data in the filtered dataset that is less than the average value of the filtered dataset ; is the data in the filtered dataset that is less than the average value of the filtered dataset ;

[0044] The upper threshold and the lower threshold in the threshold range are

[0045]

[0046] .

[0047] According to another aspect of the present invention, there is provided a parallel water pump performance monitoring device, including: a collection module, a data cleaning module, a model generation module, and an operation optimization module; wherein,

[0048] The collection module is used to monitor the parallel circulating water pump and collect the historical operation data of the parallel circulating water pump;

[0049] The data cleaning module is used to clean the collected historical operation data to obtain the repaired historical data after repair; wherein, data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair;

[0050] The model generation module is used to generate a water pump characteristic curve based on the repaired historical data and construct a parallel circulating water pump performance model;

[0051] The operation optimization module is used to monitor and optimize the operation state of the parallel circulating water pump based on the current operation data of the parallel circulating water pump by using the parallel circulating water pump performance model.

[0052] According to still another aspect of the present invention, there is provided a computing device, including: a processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0053] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to a parallel water pump performance monitoring method as described above.

[0054] According to another aspect of the present invention, a system is provided, which is capable of performing operations corresponding to a parallel water pump performance monitoring method as described above, and at least includes: a data import unit, a performance calculation unit, and a result display unit; wherein,

[0055] The data import unit is configured to import historical operation data and / or current operation data corresponding to the data import instruction into the system in response to the user's data import instruction;

[0056] The performance calculation unit is configured to use a parallel circulating water pump performance model to calculate corresponding predicted performance data according to the imported current operation data;

[0057] The result display unit is configured to display the obtained predicted performance data to the user terminal and provide a corresponding method for adjusting the operation state of the water pump.

[0058] According to the technical solution provided by the present invention, the parallel circulating water pumps are monitored, and the historical operation data of the parallel circulating water pumps are collected; the collected historical operation data are subjected to data cleaning to obtain the repaired historical data after repair; wherein, the data cleaning at least includes: abnormal data detection, abnormal data screening and / or data repair; according to the repaired historical data, a pump characteristic curve is generated, and a performance model of the parallel circulating water pump is constructed; based on the current operation data of the parallel circulating water pump, the operation state of the parallel circulating water pump is monitored and optimized by using the performance model of the parallel circulating water pump. By monitoring the parallel circulating water pumps and collecting various operation parameters corresponding to each water pump, the comprehensive monitoring and collection of the operation parameters of the parallel circulating water pumps are realized, which is beneficial to scientifically and comprehensively determining the operation conditions of the water pumps; by performing data cleaning on the collected historical operation data, for the data used for model construction, the abnormal data therein are detected and screened, and then the abnormal part is repaired based on the LSTM neural network and the SVR algorithm, greatly improving the integrity and credibility of the data, and further improving the accuracy of model construction and water pump performance prediction; based on the repaired historical operation data and the installation structure of each water pump, the local resistance coefficient is determined, and according to the calculation formula of the head, the dynamic and static heads of the water pump are more accurately determined, and thus according to the flow rate, power and speed, the corresponding characteristic curves of flow rate-head, power-head and speed-head are accurately obtained. After further determining the water pump efficiency model, water pump power model, motor efficiency model and hydraulic coupling efficiency model, polynomial fitting is performed, and accordingly a performance model of the parallel circulating water pump is scientifically constructed; based on the performance model of the parallel circulating water pump, the corresponding predicted performance data are obtained based on the current operation data, the monitoring of the parallel circulating water pump is realized, and it is compared with the monitored actual performance data. According to the results, the operation states of each water pump are scientifically adjusted, the operation parameters of the water pumps are optimized, and the operation modes of each water pump and the parallel circulating water pump as a whole are made more reasonable. While ensuring the working performance, the working efficiency of the parallel circulating water pump is greatly improved, and the overall power consumption is reduced. In addition, in the process of comparing the actual performance data with the predicted performance data, the corresponding actual performance data are determined based on the repaired historical data of a preset number of water pumps, and they are filtered according to the upper filter limit and the lower filter limit to remove invalid data. Then, by determining the average value, upper limit parameter and lower limit parameter of the filtered data set, the threshold range of the comparison result between the actual performance data and the predicted performance data is finally scientifically determined, so as to accurately judge whether the actual performance is higher or lower than the predicted data, and then reasonably determine the adjustment mode of the water pump operation.

[0059] Other features and advantages of the present invention will be set forth in the following description, and in part will be obvious from the description, or may be learned by practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the structure particularly pointed out in the written description and the drawings.

[0060] The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments. Brief Description of the Drawings

[0061] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, but do not constitute a limitation to the present invention. In the drawings:

[0062] Figure 1 A schematic flow chart of a method for monitoring the performance of parallel pumps according to an embodiment of the present invention is shown;

[0063] Figure 2 A schematic flow chart of a data cleaning method for historical operation data of parallel circulating pumps according to an embodiment of the present invention is shown;

[0064] Figure 3 A schematic flow chart of a method for constructing a performance model of parallel circulating pumps according to an embodiment of the present invention is shown;

[0065] Figure 4 A perspective view of the circulating water pipeline in a heat substation according to an embodiment of the present invention is shown;

[0066] Figure 5 An isometric view of a circulating pump and a circulating water pipeline according to an embodiment of the present invention is shown;

[0067] Figure 6 An arrangement diagram of some circulating water pipelines in a heat substation according to an embodiment of the present invention is shown;

[0068] Figure 7 A schematic diagram of Q-H curve fitting of parallel circulating pumps according to an embodiment of the present invention is shown;

[0069] Figure 8 A schematic diagram of Q-P curve fitting of parallel circulating pumps according to an embodiment of the present invention is shown;

[0070] Figure 9 A schematic diagram of Q-η curve fitting of parallel circulating pumps according to an embodiment of the present invention is shown;

[0071] Figure 10 A schematic flow chart of an operation monitoring and optimization method based on the performance model of parallel circulating pumps according to an embodiment of the present invention is shown;

[0072] Figure 11 Shows a structural block diagram of a parallel water pump performance monitoring device according to an embodiment of the present invention;

[0073] Figure 12 Shows a schematic structural diagram of a parallel water pump performance monitoring system according to an embodiment of the present invention;

[0074] Figure 13 Shows a schematic diagram of data import of a parallel water pump performance monitoring system according to an embodiment of the present invention;

[0075] Figure 14 Shows a schematic diagram of result display of a parallel water pump performance monitoring system according to an embodiment of the present invention;

[0076] Figure 15 Shows a schematic structural diagram of a computing device according to an embodiment of the present invention. Detailed implementation manners

[0077] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for explaining and illustrating the present invention, and are not used to limit the present invention.

[0078] Figure 1 Shows a schematic flowchart of a parallel water pump performance monitoring method according to an embodiment of the present invention, and the method includes the following steps:

[0079] Step S101: Monitor the parallel circulating water pump and collect the historical operation data of the parallel circulating water pump.

[0080] Preferably, the historical operation data at least includes the inlet pressure of the main pipe, the outlet pressure of each pump, the total amount of circulating water, the opening degree of each hydraulic coupling scoop tube, and / or the current of each motor.

[0081] Step S102: Perform data cleaning on the collected historical operation data to obtain the repaired historical data after repair.

[0082] Preferably, the data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair.

[0083] Furthermore, the data repair at least includes: redundant data removal, incorrect data correction, and / or missing data filling.

[0084] Step S103: Generate a water pump characteristic curve based on the repaired historical data and construct a parallel circulating water pump performance model.

[0085] Step S104: Based on the current operation data of the parallel circulating water pump, use the parallel circulating water pump performance model to monitor and optimize the operation state of the parallel circulating water pump.

[0086] According to the parallel pump performance monitoring method provided by this embodiment, the parallel circulating pumps are monitored, and the historical operation data of the parallel circulating pumps are collected; data cleaning is performed on the collected historical operation data to obtain the repaired historical data after repair; wherein, the data cleaning at least includes: abnormal data detection, data screening, and / or data repair; based on the repaired historical data, a pump characteristic curve is generated, and a parallel circulating pump performance model is constructed; based on the current operation data of the parallel circulating pump, the parallel circulating pump performance model is used to monitor and optimize the operation state of the parallel circulating pump. Through the parallel pump performance monitoring method provided by this embodiment, by monitoring the parallel circulating pump, various operation parameters corresponding to each pump are collected, realizing the comprehensive monitoring and collection of the operation parameters of the parallel circulating pump, which is beneficial to scientifically and comprehensively determining the operation condition of the pump; by performing data cleaning on the collected historical operation data, for the data used for model construction, the abnormal data therein is detected and screened, and the abnormal part is repaired, greatly improving the integrity and credibility of the data, and further improving the accuracy of model construction and pump performance prediction; based on the repaired historical operation data and the installation structure of each pump, the local resistance coefficient is determined, and according to the calculation formula of the head, the dynamic and static heads of the pump are determined more accurately, and then according to the flow rate, power, and rotational speed, the corresponding characteristic curve is accurately obtained, and then the pump efficiency model, pump power model, motor efficiency model, and hydraulic coupling efficiency model are determined, and based on this, a parallel circulating pump performance model is scientifically constructed; based on the parallel circulating pump performance model, the parallel circulating pump is monitored and its performance is predicted, and according to the comparison result of the monitored and predicted data, the operation state of each pump is scientifically adjusted, the operation parameters of the pump are optimized, making the operation mode of each pump and the parallel circulating pump as a whole more reasonable, and greatly improving the working efficiency of the parallel circulating pump while ensuring the working performance, and reducing the overall power consumption.

[0087] Figure 2 The flowchart shows a data cleaning method for the historical operation data of a parallel circulating pump according to an aspect of the present invention;

[0088] As Figure 2 shown, the method includes the following steps:

[0089] Step S201, generating structured monitoring data for the circulating pump based on the historical operation data with the same sampling frequency.

[0090] Preferably, based on the same sampling frequency, samples are taken from the historical operation data z for each timing monitoring parameter of the circulating pump to obtain the structured monitoring data of each monitoring parameter during the monitoring period.

[0091]

[0092] Among them, is the number of monitoring parameters; T is the length of the monitoring time series; is the monitoring data of the Nth monitoring parameter at time T.

[0093] Step S202, using the 3σ criterion to detect abnormal data for structured monitoring data.

[0094] Specifically, the judgment basis of the 3σ criterion for abnormal data is

[0095]

[0096] Among them, is a certain type of monitoring data; is the mean of the corresponding structured monitoring data set; is the standard deviation of the corresponding structured monitoring data set.

[0097] Step S203, based on the abnormal data detection results, screen to obtain an abnormal data set.

[0098] Preferably, in the structured monitoring data set of monitoring data, data with at least two ends of the monitoring sequences being exactly the same or overly similar are used as redundant data; data manifested as special words such as null, N / A or 0 or blanks caused by interference in sensing, transmission or storage are used as missing data; data seriously deviating from the sample distribution law or isolated points or clusters in time series are used as incorrect data.

[0099] Step S204, based on the LSTM neural network and the SVR algorithm, perform data repair on the abnormal data set.

[0100] Specifically, the data repair at least includes: redundant data removal, incorrect data correction and / or missing data filling.

[0101] Preferably, the LSTM neural network determines whether to retain data based on the sigmoid function; updates the cell state through the sigmoid function and the tanh function; after the update is completed, based on the weight matrix and bias term of the output gate, combines with the tanh function to determine the next hidden state; transfers the new cell state and the hidden state.

[0102] According to the above method, data cleaning can be performed on the collected historical operation data, so as to detect and screen out abnormal data in the data used for model construction, and then perform data repair on the abnormal part based on the LSTM neural network and the SVR algorithm, greatly improving the integrity and credibility of the data, and further improving the accuracy of model construction and pump performance prediction.

[0103] Figure 3 It shows a schematic flow chart of a method for constructing a performance model of a parallel circulating water pump according to an embodiment of the present invention;

[0104] As Figure 3 shown, the method includes the following steps:

[0105] Step S301, determine the local resistance coefficient based on the repair historical data and the structural parameters of the parallel circulating water pump, and determine the head of the parallel circulating water pump based on the dynamic head and the static head.

[0106] Specifically, the basic expression of the head is

[0107]

[0108] wherein, is the head; is the elevation head; is the density; is the acceleration of gravity; is the inlet pressure; is the outlet pressure; is the inlet flow velocity; is the outlet flow velocity; is the frictional loss.

[0109] Then the static head is

[0110]

[0111] The dynamic head is

[0112]

[0113] In the case where the pipe diameters at the inlet and outlet of the water pump are the same, the dynamic pressure difference at the inlet and outlet can be ignored, and the dynamic head is the frictional loss

[0114]

[0115] wherein, is the friction coefficient; is the pipe section length; is the pipe section diameter; is the flow velocity in the pipe; is the local resistance coefficient.

[0116] According to the structural parameters of the water pump, determine various structures included therein, such as Figure 4 、 5 、as shown in 6: Figure 4 It shows an isometric view of the circulating water pipeline of a heat station according to an embodiment of the present invention, Figure 5The isometric view of the circulating water pump and the circulating water pipeline according to an embodiment of the present invention is shown. Figure 6 The layout diagram of part of the circulating water pipeline in the heat station according to an embodiment of the present invention is shown. Further, according to the local resistance coefficient table corresponding to various structures, the corresponding local resistance coefficient is selected according to the structure parameters.

[0117] Preferably, the structures included in the water pump at least include: elbow and bend, lateral return three-way, lateral diversion three-way, combined or split fork-shaped three-way, straight pipe or lateral branch pipe of convergent or divergent inclined three-way, reducer, butterfly valve and / or check valve, etc.

[0118] For example, the local resistance coefficient table of the butterfly valve is shown as follows:

[0119]

[0120] Step S302: Based on the head of the parallel circulating water pumps, determine the corresponding flow rate, input power and rotational speed, and generate the characteristic curve of the parallel circulating water pumps.

[0121] Specifically, the characteristic curve at least includes Q-H curve, Q-P curve and Q-η curve.

[0122] Specifically, according to the proportional relationship between the target rotational speed and the rated rotational speed, combined with the flow rate, head and input power corresponding to the rated rotational speed, determine the flow rate, head and input power corresponding to the target rotational speed.

[0123] Based on the obtained flow rate Q, head H, input power P and efficiency η, obtain the corresponding Q-H curve, Q-P curve and Q-η curve;

[0124] Since the operating point of a single water pump is the intersection of the Q-H curve of the water pump and the system resistance curve, and then determine that the operating point of the parallel circulating water pumps is located at the intersection of the system resistance curve and the combined Q-H curve of all operating pumps;

[0125] Among them, for the parallel pumps composed of water pumps of the same model, based on the principle of equal head and added flow rate, obtain the Q-H curve of the parallel water pumps;

[0126] For the parallel pumps composed of water pumps of different models, taking the parallel pumps composed of two water pumps as an example, draw the Q-H curves of the two water pumps respectively, that is and ; Subsequently, based on the reduced characteristic curve method, subtract the inlet and outlet resistance loss curves of the corresponding water pumps from and to obtain the reduced characteristic curves and ; Finally, according to the principle of equal head and added flow rate, draw the The curve, i.e., the Q-H curve of a single equal-duty water pump.

[0127] Step S303, establish a water pump efficiency model, a water pump power model, a motor efficiency model, and a fluid coupling efficiency model.

[0128] Preferably, the water pump efficiency is expressed as

[0129]

[0130] Wherein, is the mechanical loss coefficient; is the friction loss coefficient; is the impact resistance loss coefficient; is the leakage loss coefficient; is the flow rate at the optimal operating point;

[0131] Accordingly, a water pump efficiency model is constructed.

[0132] The motor power factor is

[0133]

[0134] Wherein, is the motor power factor; is the motor input power; is the input voltage; is the input current;

[0135] According to the characteristic that the power factor is similar to the change rate of the motor efficiency, a motor efficiency curve is obtained, and a motor efficiency model is constructed.

[0136] The fluid coupling efficiency is

[0137]

[0138] Wherein, is the fluid coupling efficiency; is the output power of the fluid coupling; is the input power of the fluid coupling; is the volumetric efficiency; is the mechanical efficiency; is the hydraulic efficiency;

[0139] Since, in actual operation of the fluid coupling, its volumetric loss and mechanical loss are both very small and can be ignored, and their corresponding volumetric efficiency and mechanical efficiency can both be taken as 1;

[0140] And, the hydraulic efficiency can be expressed as the ratio of the turbine speed to the pump wheel speed (i.e., the ratio of the output shaft speed to the input shaft speed);

[0141] Therefore, the fluid coupling efficiency can be simplified to

[0142]

[0143] wherein, is the turbine speed; is the pump impeller speed;

[0144] Thus, the hydrodynamic coupling efficiency model is determined.

[0145] The pump power is

[0146]

[0147] Thus, the pump power model is determined.

[0148] Step S304: Perform polynomial fitting on the characteristic curves of the parallel circulating pumps, and accordingly construct the performance model of the parallel circulating pumps.

[0149] Specifically, based on the degree of the fitting polynomial, determine the fitting equation of the characteristic curve.

[0150] Preferably, taking the Q-H curve as an example, the fitting equation of the Q-H curve of the pump is

[0151]

[0152] wherein, are coefficients; is the degree of the fitting polynomial; .

[0153] The matrix form of the fitting equation is

[0154]

[0155] , ,

[0156] Solve based on the least squares method to make the fitting equation satisfy

[0157]

[0158] wherein, is the distance between the fitting curves.

[0159] Accordingly, it is obtained that

[0160] .

[0161] According to the same principle, perform polynomial fitting on other characteristic curves, such as Figure 7 , 8 , as shown in 9, Figure 7Shows the schematic diagram of the Q-H curve fitting of the parallel circulating water pump according to an embodiment of the present invention, Figure 8 Shows the schematic diagram of the Q-P curve fitting of the parallel circulating water pump according to an embodiment of the present invention, Figure 9 Shows the schematic diagram of the Q-η curve fitting of the parallel circulating water pump according to an embodiment of the present invention, which will not be elaborated here. Based on the polynomial fitting results, a performance model of the parallel circulating water pump is constructed.

[0162] According to the above method, based on the repaired historical operation data and the installation structure of each water pump, the local resistance coefficient can be determined, and according to the calculation formula of the head, the dynamic and static heads of the water pump can be more accurately determined. Then, according to the flow rate, power and speed, the corresponding characteristic curves of flow rate-head, power-head and speed-head can be accurately obtained. After further determining the water pump efficiency model, water pump power model, motor efficiency model and hydraulic coupling efficiency model, polynomial fitting is carried out, and based on this, a performance model of the parallel circulating water pump is scientifically constructed. This model can be used to accurately predict the performance of the parallel circulating water pump, so as to monitor and reasonably adjust the water pump subsequently.

[0163] Figure 10 Shows the schematic flow chart of the operation monitoring and optimization method based on the performance model of the parallel circulating water pump according to an embodiment of the present invention;

[0164] As Figure 10 shown, the method includes the following steps:

[0165] Step S1001, obtain the current operation data of the parallel circulating water pump and determine the actual performance data.

[0166] Step S1002, input the current operation data of the parallel circulating water pump into the performance model of the parallel circulating water pump, and calculate the predicted performance data.

[0167] Specifically, based on a preset number of repaired historical data, the corresponding preset number of predicted performance data are calculated as the initial data set;

[0168] Data filtering is performed on the initial data set, and the filtered data set is obtained after filtering;

[0169] Calculate the upper limit parameter and the lower limit parameter based on the filtered data set;

[0170] Based on the upper limit parameter and the lower limit parameter, determine the threshold range for comparing the actual performance data with the predicted performance data.

[0171] Preferably, based on the initial data set, calculate the average value of all the actual performance data therein , and the upper quartile and the lower quartile , and determine the upper filtering limit for filtering the initial data set accordingly and the lower filtering limit

[0172]

[0173]

[0174] According to the upper filtering limit and the lower filtering limit Filter the initial data set to obtain a filtered data set, and calculate the average value of the filtered data set

[0175]

[0176] Among them, is the number of actual performance data in the filtered data set, is the actual performance data in the filtered data set;

[0177] Upper limit parameter and lower limit parameter are

[0178]

[0179]

[0180] Among them, is the number of data in the filtered data set that is greater than the average value of the filtered data set ; is the data in the filtered data set that is greater than the average value of the filtered data set ; is the number of data in the filtered data set that is less than the average value of the filtered data set ; is the data in the filtered data set that is less than the average value of the filtered data set ;

[0181] The upper threshold and the lower threshold in the threshold range are

[0182]

[0183] .

[0184] Step S1003: Compare the actual performance data with the predicted performance data to obtain a comparison result.

[0185] Specifically, use the threshold range as the selection range of the predicted performance data, and compare it with the actual performance data accordingly.

[0186] Step S1004: Adjust the operating state of the parallel circulating water pumps according to the comparison result.

[0187] Specifically, when the actual performance data is within the threshold range, it is determined that the operating condition of the parallel circulating water pumps meets the requirements and no adjustment is required.

[0188] When the actual performance data is not within the threshold range, it is determined that the operating condition of the parallel circulating water pumps meets the requirements but the current performance is poor. Accordingly, the operating parameters of each pump are further obtained and the operating parameters of the pumps are adjusted to improve the overall performance of the parallel circulating water pumps.

[0189] Preferably, the specific adjustment method can be selected according to the actual situation and is not limited herein. For example, the adjustment method is to reduce the flow rate of the single pump with the highest power by 20% and correspondingly increase the flow rate of the pumps with power lower than the overall power of the parallel circulating water pumps by 20%.

[0190] According to the above method, based on the performance model of the parallel circulating water pumps, the corresponding predicted performance data can be obtained based on the current operating data, realizing the monitoring of the parallel circulating water pumps. Then, it is compared with the monitored actual performance data, and the operating states of each pump are scientifically adjusted according to the results, optimizing the operating parameters of the pumps, making the operating modes of each pump and the parallel circulating water pumps as a whole more reasonable. While ensuring the working performance, the working efficiency of the parallel circulating water pumps is greatly improved and the overall power consumption is reduced. In addition, during the process of comparing the actual performance data with the predicted performance data, the corresponding actual performance data is determined based on the repair historical data of a preset number of pumps, and it is filtered according to the upper and lower filtering limits to remove invalid data. Subsequently, by determining the average value, upper limit parameter, and lower limit parameter of the filtered data set, the threshold range of the comparison result between the actual performance data and the predicted performance data is finally scientifically determined, so as to accurately judge whether the actual performance is higher or lower than the predicted data, and then reasonably determine the adjustment method for the pump operation.

[0191] Figure 11 Fig. shows the structural block diagram of a parallel water pump performance monitoring device according to an embodiment of the present invention. As Figure 11 shown, the device includes: a collection module 1101, a data cleaning module 1102, a model generation module 1103, and an operation optimization module 1104; wherein,

[0192] The collection module 1101 is used to monitor the parallel circulating water pumps and collect the historical operating data of the parallel circulating water pumps.

[0193] Preferably, the historical operating data includes at least the inlet pressure of the main pipe, the outlet pressure of each pump, the total amount of circulating water, the opening of each hydraulic coupling scoop tube, and / or the current of each motor.

[0194] The data cleaning module 1102 is configured to clean the collected historical operation data to obtain the repaired historical data.

[0195] Specifically, the data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair.

[0196] Specifically, the data cleaning module 1102 is further configured to

[0197] generate structured monitoring data for the circulating water pump based on the historical operation data with the same sampling frequency;

[0198] detect abnormal data for the structured monitoring data by using the 3σ criterion;

[0199] screen out the abnormal data set based on the abnormal data detection result;

[0200] repair the abnormal data set based on the LSTM neural network and the SVR algorithm;

[0201] The data repair at least includes: redundant data removal, incorrect data correction, and / or missing data filling.

[0202] The model generation module 1103 is configured to generate a pump characteristic curve based on the repaired historical data and construct a performance model for the parallel circulating water pump.

[0203] Specifically, the model generation module 1103 is further configured to

[0204] determine the local resistance coefficient based on the repaired historical data and the structural parameters of the parallel circulating water pump, and determine the head of the parallel circulating water pump based on the dynamic head and the static head;

[0205] determine the corresponding flow rate, input power, and rotational speed based on the head of the parallel circulating water pump, and generate a characteristic curve for the parallel circulating water pump; wherein, the characteristic curve at least includes a Q-H curve, a Q-P curve, and a Q-η curve;

[0206] establish a pump efficiency model, a pump power model, a motor efficiency model, and a fluid coupling efficiency model;

[0207] perform polynomial fitting on each characteristic curve of the parallel circulating water pump, and construct a performance model for the parallel circulating water pump accordingly.

[0208] The operation optimization module 1104 is configured to monitor and optimize the operation state of the parallel circulating water pump by using the performance model of the parallel circulating water pump based on the current operation data of the parallel circulating water pump.

[0209] Specifically, the operation optimization module 1104 is further configured to

[0210] obtain the current operation data of the parallel circulating water pump and determine the actual performance data;

[0211] input the current operation data of the parallel circulating water pump into the performance model of the parallel circulating water pump, and calculate the predicted performance data;

[0212] compare the actual performance data with the predicted performance data to obtain a comparison result;

[0213] adjust the operation state of the parallel circulating water pump according to the comparison result.

[0214] Preferably, the operation optimization module 1104 is further configured to

[0215] calculate a corresponding preset number of predicted performance data based on a preset number of repair historical data as an initial data set;

[0216] perform data filtering on the initial data set, and obtain a filtered data set after completion of the filtering;

[0217] calculate an upper limit parameter and a lower limit parameter based on the filtered data set;

[0218] determine a threshold range for comparing the actual performance data with the predicted performance data based on the upper limit parameter and the lower limit parameter.

[0219] Preferably, based on the initial data set, calculate the average value of all the actual performance data therein , as well as the upper quartile and the lower quartile , and determine a filtering upper limit and a filtering lower limit

[0220]

[0221]

[0222] Filter the initial data set according to the filtering upper limit and the filtering lower limit to obtain a filtered data set, and calculate the average value of the filtered data set

[0223]

[0224] wherein, is the number of actual performance data in the filtered data set, is the actual performance data in the filtered data set;

[0225] Upper limit parameter and lower limit parameter is

[0226]

[0227]

[0228] wherein, is the number of data in the filtered dataset that is greater than the average value of the filtered dataset ; is the data in the filtered dataset that is greater than the average value of the filtered dataset ; is the number of data in the filtered dataset that is less than the average value of the filtered dataset ; is the data in the filtered dataset that is less than the average value of the filtered dataset ;

[0229] The upper threshold and the lower threshold in the threshold range are

[0230]

[0231] .

[0232] According to the parallel water pump performance monitoring device provided by this embodiment, it includes:

[0233] A collection module, a data cleaning module, a model generation module, and an operation optimization module; wherein, the collection module is used to monitor the parallel circulating water pump and collect the historical operation data of the parallel circulating water pump; the data cleaning module is used to clean the collected historical operation data to obtain the repaired historical data; wherein, data cleaning at least includes: abnormal data detection, abnormal data screening, and / or data repair; the model generation module is used to generate a pump characteristic curve based on the repaired historical data and construct a performance model of the parallel circulating water pump; the operation optimization module is used to monitor and optimize the operation state of the parallel circulating water pump based on the current operation data of the parallel circulating water pump by using the performance model of the parallel circulating water pump. Through the parallel water pump performance monitoring device provided by this embodiment, by monitoring the parallel circulating water pump and collecting various operation parameters corresponding to each water pump, the comprehensive monitoring and collection of the operation parameters of the parallel circulating water pump are realized, which is beneficial to scientifically and comprehensively determining the operation situation of the water pump; by cleaning the collected historical operation data, for the data used for model construction, the abnormal data therein is detected and screened, and then the abnormal part is repaired based on the LSTM neural network and the SVR algorithm, greatly improving the integrity and credibility of the data, and further improving the accuracy of model construction and water pump performance prediction; based on the repaired historical operation data and the installation structure of each water pump, the local resistance coefficient is determined, and according to the calculation formula of the head, the dynamic and static heads of the water pump are more accurately determined, and then based on the flow rate, power, and speed, the corresponding flow rate-head, power-head, and speed-head characteristic curves are accurately obtained, and further the water pump efficiency model, water pump power model, motor efficiency model, and hydraulic coupling efficiency model are determined, and then polynomial fitting is performed, and based on this, a performance model of the parallel circulating water pump is scientifically constructed; based on the performance model of the parallel circulating water pump, the corresponding predicted performance data is obtained based on the current operation data, realizing the monitoring of the parallel circulating water pump, and comparing it with the monitored actual performance data, and scientifically adjusting the operation state of each water pump according to the result, optimizing the operation parameters of the water pump, making the operation modes of each water pump and the parallel circulating water pump as a whole more reasonable, greatly improving the working efficiency of the parallel circulating water pump while ensuring the working performance, and reducing the overall power consumption. In addition, during the comparison process of the actual performance data and the predicted performance data, the corresponding actual performance data is determined based on the repaired historical data of a preset number of water pumps, and it is filtered according to the upper and lower filtering limits to remove invalid data, and then by determining the average value, upper limit parameter, and lower limit parameter of the filtered data set, the threshold range of the comparison result between the actual performance data and the predicted performance data is finally scientifically determined, so as to accurately judge whether the actual performance is higher or lower than the predicted data, and then reasonably determine the adjustment method of the water pump operation.

[0234] The present invention also provides a parallel pump performance monitoring system. The executable instructions of the system can execute the parallel pump performance monitoring method in any of the above method embodiments, such as Figure 12 shown Figure 12 Figure 4 shows a schematic structural diagram of a parallel pump performance monitoring system according to an embodiment of the present invention. The system includes: a data import unit 1201, a performance calculation unit 1202, and a result display unit 1203; wherein,

[0235] The data import unit 1202 is configured to import the historical operation data and / or the current operation data corresponding to the data import instruction into the system in response to the user's data import instruction.

[0236] The performance calculation unit 1202 is configured to calculate the corresponding predicted performance data according to the imported current operation data by using the parallel circulating pump performance model.

[0237] Preferably, a program written in Python language modularly is used to write the program for calculating various performance parameters of the pump in the performance calculation unit 1202.

[0238] The result display unit 1203 is configured to display the obtained predicted performance data to the user terminal and provide the corresponding method for adjusting the pump operation state.

[0239] The parallel pump performance monitoring system further includes a control unit 1204;

[0240] The control unit 1204 is configured to select the system function in response to the user's control signal.

[0241] Wherein, the user's control signal can be different signals for various terminals, such as clicks, touches, etc., which are not limited herein.

[0242] Furthermore, the styles of the parallel pump performance monitoring system during data import and result output are respectively as Figure 13 、 14 shown. Figure 13 Figure 5 shows a schematic diagram of data import of a parallel pump performance monitoring system according to an embodiment of the present invention; Figure 14 Figure 6 shows a schematic diagram of result display of a parallel pump performance monitoring system according to an embodiment of the present invention.

[0243] Figure 15 Figure 7 shows a schematic structural diagram of a computing device according to an embodiment of the present invention. The specific implementation of the computing device is not limited in the specific embodiments of the present invention.

[0244] such as Figure 15As shown in the figure, the computing device may include: a processor 1502, a communications interface 1504, a memory 1506, and a communication bus 1508.

[0245] Wherein:

[0246] The processor 1502, the communications interface 1504, and the memory 1506 communicate with each other through the communication bus 1508.

[0247] The communications interface 1504 is used to communicate with network elements of other devices such as clients or other servers.

[0248] The processor 1502 is used to execute the program 1510, and specifically can execute the relevant steps in the above-mentioned embodiments of the parallel water pump performance monitoring method.

[0249] Specifically, the program 1510 may include program code, and the program code includes computer operation instructions.

[0250] The processor 1502 may be a central processing unit (CPU), or a specific application integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the computing device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0251] The memory 1506 is used to store the program 1510. The memory 1506 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0252] The program 1510 is specifically used to cause the processor 1502 to execute a parallel water pump performance monitoring method in any of the above method embodiments. For the specific implementation of each step in the program 1510, reference may be made to the corresponding steps and units in the above-mentioned embodiments of the parallel water pump performance monitoring method, which will not be elaborated here. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding process descriptions in the foregoing method embodiments, which will not be elaborated here.

[0253] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. A variety of general-purpose systems may also be used in conjunction with the teachings presented herein. The structure required to construct such systems will be apparent from the above description. In addition, the present invention is not directed to any particular programming language. It should be understood that the teachings of the present invention described herein can be implemented in a variety of programming languages, and the description of a particular language above is for the purpose of disclosing the best mode of the present invention.

[0254] In the specification provided herein, numerous specific details are set forth. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure an understanding of the present specification.

[0255] Similarly, it should be understood that, in order to streamline the present disclosure and assist in understanding one or more of the various inventive aspects, in the foregoing description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the claims reflect, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate embodiment of the present invention.

[0256] Those skilled in the art will appreciate that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature providing the same, equivalent, or similar purpose.

[0257] In addition, those skilled in the art can understand that although some embodiments described herein include certain features included in other embodiments rather than other features, the combination of features of different embodiments means that it is within the scope of the present invention and forms different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0258] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or device program (such as a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0259] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A method for monitoring the performance of parallel pumps, comprising: Monitoring the parallel circulating pumps and collecting the historical operation data of the parallel circulating pumps; Performing data cleaning on the collected historical operation data to obtain the repaired historical data after repair; wherein, the data cleaning at least includes: abnormal data detection, abnormal data screening, and data repair; Generating a pump characteristic curve based on the repaired historical data and constructing a performance model for the parallel circulating pumps; Based on the current operation data of the parallel circulating pumps, using the performance model of the parallel circulating pumps to monitor and optimize the operation state of the parallel circulating pumps; wherein, Obtaining the current operation data of the parallel circulating pumps and determining the actual performance data; Inputting the current operation data of the parallel circulating pumps into the performance model of the parallel circulating pumps to calculate the predicted performance data; Comparing the actual performance data with the predicted performance data to obtain a comparison result; wherein, a corresponding preset number of predicted performance data are calculated based on a preset number of repaired historical data as the initial data set; data filtering is performed on the initial data set, and after completion of the filtering, a filtered data set is obtained; an upper limit parameter and a lower limit parameter are calculated based on the filtered data set; based on the upper limit parameter and the lower limit parameter, a threshold range for comparing the actual performance data with the predicted performance data is determined; Adjusting the operation state of the parallel circulating pumps according to the comparison result.

2. The method according to claim 1, wherein The monitoring of the parallel circulating pumps and the collection of the historical operation data of the parallel circulating pumps further include: The historical operation data at least includes the pressure at the inlet of the main pipe, the pressure at the outlet of each pump, the total amount of circulating water, the opening of each hydraulic coupling scoop tube, and the current of each motor.

3. The method according to claim 1, wherein The performing of data cleaning on the collected historical operation data to obtain the repaired historical data after repair; wherein, the data cleaning at least includes: abnormal data detection, abnormal data screening, and data repair, and further includes: Generating structured monitoring data for the circulating pumps based on the historical operation data with the same sampling frequency; Using the 3σ criterion to detect abnormal data for the structured monitoring data; Based on the abnormal data detection result, screening to obtain an abnormal data set; Based on the LSTM neural network and the SVR algorithm, performing data repair on the abnormal data set; The data repair at least includes: redundant data elimination, error data correction, and missing data filling.

4. The method according to claim 1, wherein The generating of a pump characteristic curve based on the repaired historical data and the constructing of a performance model for the parallel circulating pumps further include: Determining the local resistance coefficient based on the repaired historical data and the structural parameters of the parallel circulating pumps, and determining the head of the parallel circulating pumps based on the dynamic head and the static head; Based on the head of the parallel circulating pumps, determining the corresponding flow rate, input power, and rotational speed, and generating a characteristic curve of the parallel circulating pumps; wherein, the characteristic curve at least includes a Q-H curve, a Q-P curve, and a Q-η curve; Establishing a pump efficiency model, a pump power model, a motor efficiency model, and a hydraulic coupling efficiency model; Performing polynomial fitting on each characteristic curve of the parallel circulating pumps, and accordingly constructing a performance model for the parallel circulating pumps.

5. The method according to claim 1, wherein The method further includes: Based on the initial data set, calculate the average value of all prediction performance data therein , as well as the upper quartile and the lower quartile , and determine the upper filter limit and the lower filter limit According to the filtering upper limit and the filtering lower limit Filter the initial data set to obtain a filtered data set, and calculate the average value of the filtered data set Among them, is the quantity of prediction performance data in the filtered dataset, is the prediction performance data in the filtered dataset; Upper limit parameter and lower limit parameter are Among them, is the number of data in the filtered dataset that is greater than the average value of the filtered dataset ; is the data in the filtered dataset that is greater than the average value of the filtered dataset ; is the number of data in the filtered dataset that is less than the average value of the filtered dataset ; is the data in the filtered dataset that is less than the average value of the filtered dataset ; The upper limit and the lower limit of the threshold in the threshold range are 。 6. A parallel water pump performance monitoring device, comprising: An acquisition module, a data cleaning module, a model generation module, and an operation optimization module; wherein, The acquisition module is used to monitor the parallel circulation water pump and acquire the historical operation data of the parallel circulation water pump; The data cleaning module is used to clean the acquired historical operation data to obtain the repaired historical data after repair; wherein, data cleaning at least includes: abnormal data detection, abnormal data screening, and data repair; The model generation module is used to generate a water pump characteristic curve based on the repaired historical data and construct a performance model of the parallel circulation water pump; The operation optimization module is used to monitor and optimize the operation state of the parallel circulation water pump based on the current operation data of the parallel circulation water pump by using the performance model of the parallel circulation water pump; wherein, Obtain the current operation data of the parallel circulation water pump and determine the actual performance data; Input the current operation data of the parallel circulation water pump into the performance model of the parallel circulation water pump to calculate the predicted performance data; Compare the actual performance data with the predicted performance data to obtain a comparison result; wherein, a corresponding preset number of predicted performance data is calculated based on a preset number of repaired historical data as an initial data set; data filtering is performed on the initial data set, and after completion of the filtering, a filtered data set is obtained; an upper limit parameter and a lower limit parameter are calculated based on the filtered data set; based on the upper limit parameter and the lower limit parameter, a threshold range for comparing the actual performance data with the predicted performance data is determined; Adjust the operation state of the parallel circulation water pump according to the comparison result.

7. A computing device, comprising: A processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to a method for monitoring the performance of a parallel water pump according to any one of claims 1-5.

8. A parallel pump performance monitoring system, the system performs operations corresponding to a parallel pump performance monitoring method according to any one of claims 1-5, at least including: A data import unit, a performance calculation unit, and a result display unit; wherein, The data import unit is used to import the historical operation data and the current operation data corresponding to the data import instruction into the system in response to the user's data import instruction; The performance calculation unit is used to calculate the corresponding predicted performance data according to the imported current operation data by using the performance model of the parallel circulation water pump; The result display unit is used to display the obtained predicted performance data to the user terminal and provide a corresponding method for adjusting the operation state of the water pump.

Images